Systems and Methods for Determining a Diversity Penalty in Connection with Selecting Advertisement Content

2. The computer implemented method of claim 1, wherein the diversity penalty is based at least in part on a number of times the respective prospective advertisement was presented to a user of the client device.

3. The computer implemented method of claim 1, wherein the multiplier includes one of a delta entropy multiplier or a normalizing delta entropy multiplier.

4. The computer implemented method of claim 1, wherein the first entropy and the second entropy include a Shannon index.

5. The computer implemented method of claim 1, wherein the first entropy and the second entropy are each based at least in part on a number of advertisements associated with an advertiser that were previously presented to a user of the client device over a lookback period.

6. The computer implemented method of claim 5, wherein the advertisements presented to the user of the client device over the lookback period are assigned weights based on a time decay function.

7. The computer implemented method of claim 1, wherein the hypothetical benchmark advertisement includes a hypothetical advertisement that has never been presented to a user of the client device.

9. The computer-readable medium of claim 8, wherein the computer-executable instructions further include instructions that, when executed by the processor, cause the computing system to at least: determine a respective advertiser for each of the prospective advertisements, wherein the second diversity index is based at least in part on a number of advertisements associated with the respective advertiser that have been sent for presentation on the client device over a lookback period.

10. The computer-readable medium of claim 8, wherein the first diversity index and the second diversity index include a Shannon index.

11. The computer-readable medium of claim 8, wherein a lowest value for the diversity penalty is zero.

12. The computer-readable medium of claim 8, wherein the computer-executable instructions further include instructions that, when executed by the processor, cause the computing system to at least: determine a utility function to apply to the plurality of prospective advertisements; and determine the rank order based on the utility function and the diversity penalty.

13. The computer-readable medium of claim 12, wherein the utility function includes a relevance factor and a bid factor.

14. The computer-readable medium of claim 12, wherein the utility function includes a trimmer to eliminate at least one of the prospective advertisements from the plurality of advertisements prior to determining a diversity penalty for each of the plurality of prospective advertisements, wherein the trimmer eliminates the at least one of the prospective advertisements based on a ranking of the plurality of prospective advertisements determined by the utility function.

15. The computer-readable medium of claim 8, wherein the computer-executable instructions further include instructions that, when executed by the processor, cause the computing system to at least: determine that a number of advertisements presented to a user of the client device over a lookback period exceeds a threshold; and determine a normalizing factor to apply to the multiplier.

16. The computer-readable medium of claim 8, wherein the computer-executable instructions further include instructions that, when executed by the processor, cause the computing system to at least: determine that a number of advertisements presented to a user of the client device over a lookback period is below a threshold; and determine a delta entropy multiplier to apply as the multiplier.

18. The computing system of claim 17, wherein the multiplier is based at least in part on a number of advertisements presented to a user of the client device during a lookback period.